esm: unflag --experimental-wasm-modules

[guybedford]

This unflags --experimental-wasm-modules for Node.js 24, while keeping the implementation experimental for now.

With #56919 landed, Node.js supports both source phase imports and instance phase imports to WebAssembly modules and for Wasm imports to JS, in line with the current Phase 3 WebAssembly ESM Integration proposal (https://github.com/webassembly/esm-integration).

By unflagging, this will enable build tools to target source phase imports for Node.js as an output format, providing immediate benefit to Wasm consumers in enabling more seamless interoperability between JS and Wasm modules.

[nodejs-github-bot]

Review requested:

• @nodejs/loaders

[guybedford]

//cc @nodejs/loaders

[marco-ippolito]

So I assume its semver major?

[guybedford]

Strictly speaking, this isn't semver major since it has no breaking semantics. But, we only have fully V8 support on 24 unless there is a backport of the V8 update.

[codecov]

Codecov Report

Attention: Patch coverage is 66.66667% with 1 line in your changes missing coverage. Please review.

Project coverage is 89.10%. Comparing base (79f96b6) to head (b72069f).
Report is 310 commits behind head on main.

Files with missing lines	Patch %	Lines
lib/internal/modules/esm/formats.js	50.00%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #57038      +/-   ##
==========================================
- Coverage   89.10%   89.10%   -0.01%     
==========================================
  Files         665      665              
  Lines      193203   193196       -7     
  Branches    37220    37215       -5     
==========================================
- Hits       172158   172146      -12     
+ Misses      13771    13767       -4     
- Partials     7274     7283       +9     

Files with missing lines	Coverage Δ
lib/internal/modules/esm/translators.js	91.42% <ø> (-0.04%)	⬇️
src/node_options.cc	87.97% <100.00%> (ø)
lib/internal/modules/esm/formats.js	98.55% <50.00%> (-0.10%)	⬇️

... and 32 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[mcollina]

lgtm

[github-actions]

The notable-change PRs with changes that should be highlighted in changelogs. label has been added by @mcollina.

Please suggest a text for the release notes if you'd like to include a more detailed summary, then proceed to update the PR description with the text or a link to the notable change suggested text comment. Otherwise, the commit will be placed in the Other Notable Changes section.

[guybedford]

I've also added a commit here to remove the experimental warning, given we are on a stabilization path for this feature, but for those who have already approved feel free to comment further - @mcollina @anonrig.

[legendecas]

I think "unflag an experimental feature" should start with switching the default value to be true, and allowing turning the feature off with --no-experiemental-wasm-modules, instead of making it a no-op directly.

BTW, if we are going to make this a no-op, EnvironmentOptions::experimental_wasm_modules should be removed as it is not used.

[aduh95]

I agree, if having WASM module enabled breaks something, it's useful to be able to disable it. We can make the falg a no-op in another major

[GeoffreyBooth]

Suggested change

	> Stability: 1 - Experimental
	> Stability: 1.2 - Release candidate

[guybedford]

Thanks for all the approvals it is great to see the interest in this.

After further discussion, since this is still a Phase 3 Wasm specification, that implies it is still in experimental implementation status until it reaches Phase 4.

While strictly speaking we can land experimental features unflagged in Node.js, the ESM Integraiton has already taken long enough that there is no need to rush something out the door here - so I will hold off on landing this for Node.js 24 and instead seek to land it for (hopefully) 25 instead where ideally we would be able to land under Phase 4 later in the year, or to then at least reconsider unflagging under Phase 3 if we feel it is in the interests of the project at that point for our users.

[marco-ippolito]

We discussed at the collab summit and we realized its not semver major so it can land

[guybedford]

Thanks for clarifying the case here, I'll aim to land this then when the current design discussions on the ESM Integration proposal repo have been resolved (theres a few final PRs in progress and under discussion right now - https://github.com/WebAssembly/esm-integration/pulls). Hopefully these should resolve fairly soon, but it may be more than a few weeks yet too.